Fwd: Insides of a Widget disk

<x-html><!x-stuff-for-pete base="" src="" id="0" charset="UTF-8"><div dir="ltr"><br><br><div class="gmail_quote">---------- Forwarded message ----------<br>From: <b class="gmail_sendername">PSchaefer</b> <span dir="ltr">&lt;<a href="mailto:dr.p.schaefer_at_email.domain.hidden">dr.p.schaefer_at_email.domain.hidden</a>&gt;</span><br>

Date: Thu, Apr 24, 2014 at 8:03 AM<br>Subject: Re: Insides of a Widget disk<br>To: <a href="mailto:uo957_at_email.domain.hidden">uo957_at_email.domain.hidden</a><br><br><br><div dir="ltr">Am Freitag, 31. Januar 2014 19:30:36 UTC+1 schrieb compu_85:<br>

<br>&gt; Several years ago I picked up an &quot;as is&quot; Widget disk from eBay. I didn&#39;t pay much, and<br>&gt; wasn&#39;t upset when it made horrible grinding noises when powered up. <br><br>Most noise comes from the spindle motor bearing. Sometimes it helps running the drive upside down for a few hours. Remove the Z8 chip from the controller board, turn the drive upside down, cover it with a newspaper to keep it warm, and leave it running overnight. This rearranges the grease inside the bearings. The drive must not be operated in this position -- remove the controller chip to tie /PReset down to keep it inactive!<br>

<br>&gt; Last night I was fiddling with my Lisa again and recalled a post saying there was a glass<br>&gt; graticule inside the drive. If you look closely you will see a corner is chipped off of the glass.<br>&gt; I believe that is why this drive failed... glass chunks got under the lower head and it scratched<br>

&gt; all the disk surface off. <div><br>The ceramic heads are much more robust than the aluminium platters. Therefore you can try to replace the magnetic disk.<br><br>Loosen the outer endstop (the screw with rubber ring right next to the actuator arm. Get a small piece of plastic material (Teflon preferred), about 1x1 cm2. Slide the heads from the disk over the plastic. They should not touch themselves, due to their polished surface they will stick together. Loosen the three UNC 6-32 screws at the spindle and remove the platter. Now take your vacuum cleaner and remove any debris.<br>

<br>You can get a spare disk from an old MFM drive. It needs to have the same orange color as the original one, black (CrO2) or silver (pure iron) surfaces will not work. I have successfully used platters from a NEC D5126. To slide the heads onto the disk, tilt the plastic to move them a few millimeters apart. After reassembly, format the drive 2-3 times and run a surface scanner (e.g. BLU).<br>

<br>If you do this on your bench, the drive will fail again after a few weeks (after collecting more and more spares) because dust particles will scratch the surface. The absolute worst-case is a dust particle caught between head and disk -- at 2700 rpm this will polish away the magnetic coating! However, if done properly inside a class 100 cleanroom, reliable results are possible.<br>

<br><br>There is another issue that needs to be resolved: Widgets servo system uses two steps to approach a track on the disk. At first, coarse positioning is done with the A/B signals from the glass scale ( view testpoint TP9 + 8 in X-Y mode). Second, the heads are adjusted for the maximum signal from the RW heads (Pos Error at TP3). There is a DAC on the servo board that generates this ATF track offset signal.<br>

<br>As there is no was to completely erase the magnetic surfaces, during the formatting process the new tracks are centered onto the old ones. This is a good idea from crosstalk point of view, but it will cause trouble if the old tracks are at the edge of the ATF control range. Apple&#39;s formatter routines reject any track with an ATF offest above +/-15 (+/-31 is limit). With some unknown data on a recycled disk, the ATF will lock at +31, i.e. it will not work at all. Therefore you can expect an increased soft read error rate (08 00 00 C1) from such a drive. <br>

<br>My formatting utility gives a warning if the ATF offset is above +/-16, but continues formatting. I recommend that BLU should to the same. Being too strict at this point would ban too many drives from being repaired. <br>

<br>The correct solution would be to completely degauss the disk (apply a 4f signal to the heads and move in 1/4 track increments), then force the ATF into zero position (read spare table into RAM, set offset table to 00, write back), then write brandnew tracks into the new &#39;optimum&#39; positions, and then repeat this with ATF enabled. I think the first step is not possible with the original R/W board, it will need special (FST) hardware.<br>

<br><br>Regards<br>Patrick<span class="HOEnZb"><font color="#888888"><br><br><br><br><br></font></span></div></div><span class="HOEnZb"><font color="#888888">

<p></p>

• <br>
• <br> -----<br> You received this message because you are a member of the LisaList group.<br> The group FAQ is at <a href="http://lowendmac.com/lists/lisa.html" target="_blank">http://lowendmac.com/lists/lisa.html</a><br> To post to this group, send email to <a href="mailto:lisalist_at_email.domain.hidden" target="_blank">lisalist_at_email.domain.hidden</a><br> To leave this group, send email to <a href="mailto:lisalist%2Bunsubscribe_at_email.domain.hidden" target="_blank">lisalist+unsubscribe_at_email.domain.hidden</a><br> For more options, visit this group at <a href="http://groups.google.com/group/lisalist" target="_blank">http://groups.google.com/group/lisalist</a><br> --- <br> You received this message because you are subscribed to the Google Groups &quot;LisaList&quot; group.<br> To unsubscribe from this group and stop receiving emails from it, send an email to <a href="mailto:lisalist+unsubscribe_at_email.domain.hidden" target="_blank">lisalist+unsubscribe_at_email.domain.hidden</a>.<br> For more options, visit <a href="https://groups.google.com/d/optout" target="_blank">https://groups.google.com/d/optout</a>.<br>
  </font></span></div><br></div>
  

</x-html>
Received on 2014-06-13 13:10:12